Method and system for detecting illegal advertising pictures based on deep learning target detection
A target detection and deep learning technology, applied in neural learning methods, biological neural network models, calculations, etc., can solve problems such as corner point matching confusion
Active Publication Date: 2022-02-18
万商云集(成都)科技股份有限公司
View PDF24 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The current method of detecting illegal pictures based on deep learning mainly predicts an offset value through each corner point in the upper left corner and lower right corner of the picture. If an offset value in the upper left corner is similar to an offset value in the lower right corner, the Two points are matched into a box, which will lead to confusion in corner matching, especially when two similar objects are relatively close, often matching the upper left and lower right corners that do not belong to the same target Together, therefore, it is necessary to provide a solution to improve the accuracy of illegal picture detection results while improving detection efficiency
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0034] The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.
[0035] It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.
[0036] Please see figure 1 , figure 1 It is a schematic flowchart of a method for detecting illegal advertisement pictures based on deep learning target detection provided by an embodiment of the present invention.
[0037]According to the applicant's research, the current method of detecting illegal pictures based on deep learning mainly predicts an offset value throu...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention provides a method and system for detecting illegal advertisement pictures based on deep learning target detection. The prediction frame composed of points analyzes whether the center point of the prediction frame is within the preset range near the center point of the real target frame; if the center point of the prediction frame is within the preset range near the center point of the real target frame, the prediction The frame is divided into a valid prediction frame, otherwise the prediction frame is converted into an invalid prediction frame; then, a RoIAlign layer is used to analyze the matching value between the upper left corner point and the lower right corner point of the valid prediction and invalid prediction frame respectively, and according to The matching value constructs a loss function to optimize the recognition result; finally, the effective prediction box whose optimized matching value is greater than the set threshold is output as the output result.
Description
technical field [0001] The present invention relates to the technical field of illegal picture detection, in particular to a method and system for detecting illegal advertising pictures based on deep learning target detection. Background technique [0002] The current review of illegal content often adopts the form of manual review. The reviewers analyze and judge the content information one by one, and it is difficult to guarantee both efficiency and accuracy. Today, as technology continues to mature, artificial intelligence technologies such as natural language processing, image recognition, and voiceprint recognition have been used in some fields. The introduction of artificial intelligence technology can completely change the traditional form of content review and realize real-time review of Internet content information. Both audit efficiency and audit accuracy will be greatly improved. The current method of detecting illegal pictures based on deep learning mainly pred...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06V10/24G06N3/08
CPCG06N3/08
Inventor 王飞田文洪马霆松宋净安
Owner 万商云集(成都)科技股份有限公司